Is this you? Claim this profile to correct it, add a bio and choose the work people see first.
Claim this profileWorks229 from public data
- Pfp535
Through empirical study on a large dataset of 802,939 Web pages and 1,021,107 tags, it is demonstrated that PFP can achieve virtually linear speedup and to be promising for supporting query recommendation for search engines.
- Combating Fake News509
This survey describes the modern-day problem of fake news and, in particular, highlights the technical challenges associated with it and comprehensively compile and summarize characteristic features of available datasets.
- Session-Based Social Recommendation via Dynamic Graph Attention Networks488
This work proposes a recommender system for online communities based on a dynamic-graph-attention neural network, which dynamically infers the influencers based on users' current interests and can be efficiently fit on large-scale data.
- Topic sentiment analysis in twitter484
This study focuses on hashtag-level sentiment classification, which aims to automatically generate the overall sentiment polarity for a given hashtag in a certain time period, and proposes a novel graph model and three approximate collective classification algorithms for inference.
- Understanding the Limiting Factors of Topic Modeling via Posterior Contraction Analysis284
This paper presents theorems elucidating the posterior contraction rates of the topics as the amount of data increases, and a thorough supporting empirical study using synthetic and real data sets, including news and web-based articles and tweet messages.
- We know what @you #tag266
This work proposes comprehensive measures to quantify the major factors of how a user selects content tags as well as joins communities, and proves the effectiveness of the dual role, where both the content measures and the community measures significantly correlate to hashtag adoption on Twitter.
- DisenHAN: Disentangled Heterogeneous Graph Attention Network for Recommendation188
This paper uses meta relations to decompose high-order connectivity between node pairs and proposes a disentangled embedding propagation layer which can iteratively identify the major aspect of meta relations in a heterogeneous information network.
- An Attention-based Collaboration Framework for Multi-View Network Representation Learning181
Experimental results on real-world networks show that the proposed approach outperforms existing state-of-the-art approaches for network representation learning with a single view and other competitive approaches with multiple views.
- 161
- A Survey of Reasoning with Foundation Models: Concepts, Methodologies, and Outlook142
This article introduces seminal foundation models proposed or adaptable for reasoning, highlighting the latest advancements in various reasoning tasks, methods, and benchmarks, and delves into the potential future directions behind the emergence of reasoning abilities within foundation models.
- An Ensemble of Retrieval-Based and Generation-Based Human-Computer Conversation Systems112
Experimental results show that a novel ensemble of retrieval-based and generation-based conversation system outperforms each single module by a large margin.
- A Diffusion Model for POI Recommendation107
Inspired by the wide application of diffusion algorithm in sampling from distributions, Diff-POI encodes the user’s visiting sequence and spatial character with two tailor-designed graph encoding modules, followed by a diffusion-based sampling strategy to explore the user’s spatial visiting trends.
- DisenPOI: Disentangling Sequential and Geographical Influence for Point-of-Interest Recommendation104
This paper proposes DisenPOI, a novel Disentangled dual-graph framework for POI recommendation, which jointly utilizes sequential and geographical relationships on two separate graphs and disentangles the two influences with self-supervision.
- 103
- Text Classification with Heterogeneous Information Network Kernels98
A novel text as network classification framework, which introduces a structured and typed heterogeneous information networks (HINs) representation of texts, and a meta-path based approach to link texts that outperforms the state-of-the-art methods and other HIN-kernels.
- Few-shot Molecular Property Prediction via Hierarchically Structured Learning on Relation Graphs84
A novel framework called Hierarchically Structured Learning on Relation Graphs (HSL-RG) for molecular property prediction, which explores the structural semantics of a molecule from both global-level and local-level granularities, and proposes a task-adaptive meta-learning algorithm to provide meta knowledge customization for different tasks in few-shot scenarios.
- Learning Graph ODE for Continuous-Time Sequential Recommendation78
A novel ordinary differential equation based GNN is introduced to implicitly model the temporal evolution of the user-item interaction graph and an attention-based GNN is proposed to explicitly incorporate collaborative attention to interaction signals when the interaction graph evolves over time.
- 73
- Self-supervised Graph-level Representation Learning with Adversarial Contrastive Learning71
This work proposes a Graph Adversarial Contrastive Learning (GraphACL) scheme that learns a bank of negative samples for effective self-supervised whole-graph representation learning and optimize the parameters of the graph encoding branch and adversarial generation branch alternately.
- Dynamic Hypergraph Structure Learning for Traffic Flow Forecasting71
This paper proposes a novel model named Dynamic Hypergraph Structure Learning (DyHSL) for traffic flow prediction, which extracts hypergraph structural information to model dynamics in the traffic networks, and updates each node representation by aggregating messages from its associated hyperedges.
- 66
- Unsupervised graph-level representation learning with hierarchical contrasts66
An unsupervised graph-level representation learning framework named Hierarchical Graph Contrastive Learning (HGCL), which investigates the hierarchical structural semantics of a graph at both node and graph levels and significantly outperforms a broad range of state-of-the-art baselines.
- 64
- An agent-oriented approach to resolve scheduling optimization in intelligent manufacturing64
The architecture of intelligent manufacturing, the scheduling optimization algorithm, the negotiation processes and protocols among the agents are described in detail and a prototype system is built and validated in an illustrative example, which demonstrates the feasibility of the proposed approach.
- SUIT: A Supervised User-Item Based Topic Model for Sentiment Analysis62
A new Supervised User-Item based Topic model, called SUIT model, is proposed, which can simultaneously utilize the textual topic and latent user-item factors and shows significant improvement compared with supervised topic models and collaborative filtering methods.
- 60
- KnowSim: A Document Similarity Measure on Structured Heterogeneous Information Networks57
This work proposes a method to represent a document as a typed heterogeneous information network (HIN), where the entities and relations are annotated with types, and proposes to use the meta-path defined in HIN to compute distance between documents.
- 57
- Poisoning medical knowledge using large language models55
Scorpius, a malicious text generator, is developed to investigate whether large language models can mislead medical knowledge graphs, and shows that a single generated paper abstract can mislead a medical reasoning system that has read millions of papers.
- Learning on Graphs under Label Noise55
This work investigates the problem of learning on graphs with label noise and develops a novel approach dubbed Consistent Graph Neural Network (CGNN), which promotes two views of augmented nodes to have consistent representations and can enhance the robustness of node representations to label noise.
- 55
- One theme in all views55
A general solution that is able to exploit multiple types of contexts without arbitrary manipulation of the structure of classical topic models is proposed, and a co-regularization framework is proposed to let these views collaborate with each other, vote for the consensus topics, and distinguish them from view-specific topics.
- Web Service Composition Using Markov Decision Processes55
A method for dynamic web service composition, which is based on Markov Decision Processes (MDP), is defined on the base of QoS description and addresses the issue of selecting web services for the purpose of their composition.
- 52
- Hypergraph-enhanced Dual Semi-supervised Graph Classification50
A Hypergraph-Enhanced DuAL framework named HEAL for semi-supervised graph classification is proposed, which captures graph semantics from the perspective of the hypergraph and the line graph, respectively, and designs a hypergraph structure learning to adaptively learn complex node dependencies beyond pairwise relations.
- DisenSemi: Semi-Supervised Graph Classification via Disentangled Representation Learning50
A novel framework termed DisenSemi is introduced, which learns disentangled representation for semi-supervised graph classification and defines an MI-based disentangled consistency regularization between two models and identifies the corresponding rationale that aligns well with the current graph classification task.
- Towards Semi-Supervised Universal Graph Classification49
This paper proposes a novel graph neural network framework named UGNN, which makes the best of unlabeled data from the subgraph perspective and estimates the certainty of unlobaled graphs using multiple subgraphs, which facilities the discovery of unlabelled data from unknown categories.
- Redundancy-Free Self-Supervised Relational Learning for Graph Clustering49
This article proposes a novel self-supervised deep graph clustering method named relational redundancy-free graph clustering (R2FGC), which extracts the attribute- and structure-level relational information from both global and local views based on an autoencoder (AE) and a graph AE (GAE).
- A role of Rab7 in stabilizing EGFR‐Her2 and in sustaining Akt survival signal49
Results suggest that Rab7 is likely involved in protecting EGFR and Her2 from being degraded by the proteosome and in maintaining optimal Akt survival signal (especially during cell detachment or when HSP90 is inhibited) and is potentially a novel target for combinatory therapy with Hsp90 inhibitors.
- Learning the Joint Representation of Heterogeneous Temporal Events for Clinical Endpoint Prediction47
A novel model for learning the joint representation of heterogeneous temporal events is proposed which adds a new gate to control the visiting rates of different events which effectively models the irregular patterns of different Events and their nonlinear correlations.
- TGNN: A Joint Semi-supervised Framework for Graph-level Classification46
This paper proposes a novel semi-supervised graph classification framework called Twin Graph Neural Network (TGNN), which has two twin modules that collaborate with each other by exchanging instance similarity knowledge to fully explore the structure information of both labeled and unlabeled data.
- GHNN: Graph Harmonic Neural Networks for semi-supervised graph-level classification46
This paper proposes Graph Harmonic Neural Network (GHNN), which combines the advantages of both worlds to sufficiently leverage the unlabeled data, and thus overcomes label scarcity in semi-supervised scenarios.
- KGNN45
This paper proposes the Kernel-based Graph Neural Network (KGNN), a network that consists of a GNN-based network as well as a kernel- based network parameterized by a memory network, and jointly train the two networks by maximizing their agreement on unlabeled graphs via posterior regularization.
- A bioactivity foundation model using pairwise meta-learning44
Promising results indicate that ActFound could be an effective bioactivity foundation model for compound bioactivity prediction, paving the way for machine-learning-based drug development and discovery.
- DisenCite: Graph-Based Disentangled Representation Learning for Context-Specific Citation Generation42
A novel disentangled representation based model DisenCite is proposed to automatically generate the citation text through integrating paper text and citation graph, empowering the generation of different types of citations for the same paper.
- 41
- Question answering based on pervasive agent ontology and Semantic Web40
This paper presents a QA system in which the domain knowledge is represented by means of ontology, and personalized services are enabled through modeling users' profiles in the form of pervasive agent ontology.
- 39
- Dialogue Session Segmentation by Embedding-Enhanced TextTiling39
This paper aims to demonstrate the efforts towards in-situ applicability of EMMARM, the objective of which is to provide real-time information about concrete mechanical properties such as E-modulus and compressive strength.
- GALA: Graph Diffusion-Based Alignment With Jigsaw for Source-Free Domain Adaptation38
This study proposes a novel method named Graph Diffusion-based Alignment with Jigsaw (GALA), tailored for source-free graph domain adaptation, and develops a simple yet effective graph-mixing strategy named graph jigsaw to combine confident graphs and unconfident graphs, which can enhance generalization capabilities and robustness via consistency learning.
- DisenCTR38
The key novelty of the method compared with existing approaches is to model evolving diverse interests of users, and based on the rich dynamics supplied by the graph, a disentangled graph representation module is proposed to extract diverse user interests.
- Ultrafast imaging of spontaneous symmetry breaking in a photoionized molecular system38
It is unraveled that the Jahn-Teller deformation from C3v to C2v geometries takes a characteristic time of 20 ± 7 fs for this system, and classical and quantum molecular dynamics simulations agree well with the measurement.
- “Small Dots, Multiple Functions”: Unveiling the Veil of a Bioactive Quercetin Carbon Dot and Its Multifaceted Antibacterial and Osteogenesis Mechanisms for Infectious Bone Defect Repair37
The chemical structure of QCDs is revealed, and it is found they can interact with RTKs and thus activate the Rap1 signaling pathway to promote bone regeneration and kill bacteria by destroying their walls and interfering with energy metabolism, thus providing insights for developing a simplified component with multiple functions for repairing infectious bone defects.
- Toward Effective Semi-supervised Node Classification with Hybrid Curriculum Pseudo-labeling36
This work develops a novel framework termed Hybrid Curriculum Pseudo-Labeling (HCPL) for efficient semi-supervised node classification that incorporates both prediction confidence and uncertainty under random perturbations, therefore mitigating the influence of erroneous pseudo-labels.
- Metadata Extraction from Bibliographies Using Bigram HMM36
Experiments show that the bigram HMM yields the best result and seem to be the most promising candidate for metadata extraction of bibliographies.
- Kernel-based Substructure Exploration for Next POI Recommendation34
A Kernel-Based Graph Neural Network (KBGNN) is proposed for next POI recommendation, which combines the characteristics of both geographical and sequential influences in a collaborative way and demonstrates the superior performance of the proposed method over the state-of-the-arts.
- A trigram hidden Markov model for metadata extraction from heterogeneous references34
The results showed overall accuracy, precision, recall and F1 measure of over 95% suggesting that the method outperforms other related methods in the task of metadata extraction from references.
- 33
- Learning Hybrid Representations for Automatic 3D Vessel Centerline Extraction33
This work proposes a hybrid representation learning approach to address the challenge of preserving the continuity of extracted vessels when segmenting such thin tubular structures from 3D images, and validates the proposed approach on CTA datasets and demonstrates its superior performance compared to both traditional and CNN-based baselines.
- 33
- 33
- Participation Maximization Based on Social Influence in Online Discussion Forums32
This paper proposes a heuristic algorithm, named Thread Allocation Based on Influence (TABI), which consistently outperforms all other algorithms in maximizing participation and demonstrates that current recommender systems can be made more effective by considering future influence propagations.
- Integrating Temporal Usage Pattern into Personalized Tag Prediction32
The temporal factor in users' tagging behaviors is examined by investigating the occurrence patterns of tags and then incorporated into a novel method for ranking tags, which outperforms other temporal models and state-of-the-art tag prediction methods.
- RAHNet: Retrieval Augmented Hybrid Network for Long-tailed Graph Classification31
A novel framework called Retrieval Augmented Hybrid Network (RAHNet) is proposed to jointly learn a robust feature extractor and an unbiased classifier in a decoupled manner and innovatively optimize a category-centered supervised contrastive loss to obtain discriminative representations, which is more suitable for long-tailed scenarios.
- 30
- Towards a Neural Conversation Model With Diversity Net Using Determinantal Point Processes30
A novel framework which organically combines seq2seq model with Determinantal Point Processes is proposed, which achieves high quality in generated reply and significantly improves the diversity among them.
- Opportunities or risks to reduce labor in crowdsourcing translation? characterizing cost versus quality via a pagerank-HITS hybrid model30
This paper proposes a graphbased PageRank-HITS Hybrid model to distinguish authoritative workers from unreliable ones, and demonstrates the algorithm will keep the performance while reduce work force and hence cut cost.
- A Paper Recommender for Scientific Literatures Based on Semantic Concept Similarity29
A recommender for scientific literatures based on semantic concept similarity computed from the collaborative tags is presented, and content-based filtering approach is used to generate recommendation list from the papers these neighbor users tagged.
- Undergraduate computer science education in China27
The evolution of computing education over the past decade and its role in meeting society's requirements for IT talent, and developing trends for computing education at China's universities are described.
- Unsupervised meta-path selection for text similarity measure based on heterogeneous information networks25
This paper proposes to use an approximate personalized PageRank algorithm to find useful subgraphs to allocate the meta-paths, and develops a new similarity measure called KnowSim which is an ensemble of selected meta- PathSim, which results in impressive high-quality document clustering and classification performance.
- ALEX: Towards Effective Graph Transfer Learning with Noisy Labels24
A novel technique termed Balance Alignment and Information-aware Examination (ALEX) is introduced to address the problem of graph transfer learning in the presence of label noise, which transfers knowledge from a noisy source graph to an unlabeled target graph.
- 23
- 22
- 22
- Distant Meta-Path Similarities for Text-Based Heterogeneous Information Networks22
The distant meta-path similarity that is able to capture HIN semantics between two distant (isolated) entities to provide more meaningful entity proximity is proposed and the state-of-the-art similarity performance on two text-based HINs is shown.
- 21
- Towards Long-Tailed Recognition for Graph Classification via Collaborative Experts20
To equilibrate the contributions of head and tail classes, a novel long-tailed graph-level classification framework via Collaborative Multi-expert Learning (CoMe) is proposed, which first develops balanced contrastive learning from the view of representation learning, and then design an individual-ex Expert classifier training based on hard class mining.
- Deep Graph Mutual Learning for Cross-domain Recommendation20
This paper proposes a novel Deep Graph Mutual Learning framework (DGML) for cross-domain recommendation that first separately construct domain-shared and domain-specific interaction graphs, and develops a parallel graph neural network to extract user preference in corresponding graph.
- Please spread: recommending tweets for retweeting with implicit feedback20
A novel retweet recommendation (RTR) framework which leverages the implicit feedback to help user find the potential tweets he may want to retweet so that tweets can be taken into account as soon as it is published.
- 20
- A Linear Text Classification Algorithm Based on Category Relevance Factors20
Experimental results show that CRF outperforms k-NN and is competitive with SVM.
- Rank and Align: Towards Effective Source-free Graph Domain Adaptation19
This work investigates an underexplored yet practical problem of source-free graph domain adaptation, which transfers knowledge from source models instead of source graphs to a target domain, and introduces a novel GNN-based approach called Rank and Align (RNA), which ranks graph similarities with spectral seriation for robust semantics learning.
- Learning Knowledge-diverse Experts for Long-tailed Graph Classification18
The Knowledge-diverse Experts (KDEX) for long-tailed graph classification introduces a knowledge-diverse expert training approach to train experts with different capacities in managing various test distributions and outperforms current baselines in both standard and test-agnostic long-tailed graph classification.
- 18
- 18
- 17
- Collaborative Users’ Brand Preference Mining across Multiple Domains from Implicit Feedbacks17
A latent factor model to collaboratively mine users’ brand preferences across multiple domains simultaneously by collective learning so that the problem of data scarcity in each single domain can be effectively addressed.
- Recommending Scientific Literatures in a Collaborative Tagging Environment17
A tag-based recommendation framework for scientific literatures which models the user interests with tags and literature keywords and is better than the baseline algorithm and the extension of user model and dot-product-based similarity computation are also helpful.
- Poisoning scientific knowledge using large language models15
Scorpius is developed, a conditional text generation model that generates a malicious paper abstract conditioned on a promoting drug and a target disease that demonstrates the possibility of poisoning scientific knowledge graphs and manipulating downstream applications using LLMs.
- 15
- User browsing behavior-driven web crawling15
The proposed URL pattern-based crawl orderings are capable to properly predict the importance of newly created (unseen) URLs and have the potential to improve the performance of web crawlers.
- AlphaEval: A Comprehensive and Efficient Evaluation Framework for Formula Alpha Mining14
AlphaEval is proposed, a unified, parallelizable, and backtest-free evaluation framework for automated alpha mining models that effectively identifies superior alphas compared to traditional single-metric screening approaches.
- Socialized Word Embeddings14
This paper proposes a socialized word embedding al- algorithm which can consider both user’s personal characteristics of language use and the user's social relationship on social media, and uses a user vector to represent each user.
- Multiagent-based scheduling optimization for Intelligent Manufacturing System14
Experiments do prove that it is feasible to use the Multiagent-based scheduling method to develop an Intelligent Manufacturing System, which is valuable for further study in more depth.
- 13
- Criterion-based Heterogeneous Collaborative Filtering for Multi-behavior Implicit Recommendation13
A novel non-sampling learning framework named Criterion-guided Heterogeneous Collaborative Filtering (CHCF), which integrates criterion learning and user preference learning into a unified framework, which can be trained jointly for the interaction prediction of the target behavior.
- Early Prediction of Sepsis From Clinical Data via Heterogeneous Event Aggregation13
The proposed Heterogeneous Event Aggregation can not only shorten the length of clinical event sequence but also help to retain temporal interactions of both categorical and numerical features of clinical events in the multiple heads of the aggregation representations.
- Diversifying Neural Conversation Model with Maximal Marginal Relevance.13
This paper proposes the MMR-BS decoding method, which incorporates MMR into the beam search (BS) process of seq2seq, and shows that the proposed model achieves the best performance among other comparison methods.
- SVM+BiHMM: A Hybrid Statistic Model for Metadata Extraction13
The complementary SVM+BiHMM model is built based on the BiH MM model, with the emitting probability adjusted by the Sigmoid function of SVM score, and the transition probability trained by Bigram HMM.
- A Fast and Accurate Method for Approximate String Search13
Experimental results on large scale data show that the proposed approach improves upon existing methods in terms of accuracy in different settings.
- 13
- 12
- Multinational Perspectives on Information Technology from Academia and Industry12
With the IT Model Curricula of CC2005, IT2008, IT2017, extensive literature review, and the multinational insights of the authors into the status of IT, this paper presents a comprehensive overview and discussion of future directions of global IT education toward 2025.
- 12
- Learning to rank audience for behavioral targeting in display ads12
A novel probabilistic generative model named Rank Latent Dirichlet Allocation (RANKLDA) to rank audience according to their ads click probabilities for the long tail advertisers to deliver their ads.
- Step-by-Step Regression: A More Efficient Alternative for Polynomial Multiple Linear Regression in Stream Cube12
Step-by-step Regression not only significantly alleviates the computation pressure on the front ends of data stream management systems, but also results in a much smaller stream cube for on line analysis and real-time surveillance.
- 10
- PolyCF: Towards Optimal Spectral Graph Filters for Collaborative Filtering9
PolyCF is proposed, a flexible graph signal filter that leverages polynomial graph filters to process interaction signals and is able to approximate the optimal polynomial response function for recovering missing interactions.
- 9
- 9
- 9
- Privacy and security in recommenders: an analytical review8
A comprehensive and systematic review of the underlying causes of privacy and security challenges in RS is presented and a detailed taxonomy categorizing these concerns based on their targets and the risks they create is provided.
- Supervised topic model with consideration of user and item8
A new supervised topic model by incorporating the user and the item information is proposed that can simultaneously utilize the textual topic and user-item factors for label prediction and shows clear improvement compared with traditional supervised topic models and recommendation method.
- Web-based multimedia information retrieval for clinical application research8
A web-based data warehousing method for retrieving and analyzing neurological multimedia information for epilepsy research that supports convenient access, effective search and retrieval of clinical textual and image data, and on-line analysis.
- Cross-Domain Diffusion With Progressive Alignment for Efficient Adaptive Retrieval7
This approach revisits unsupervised efficient domain adaptive retrieval from a graph diffusion perspective, simulating cross-domain adaptation dynamics to achieve a stable target domain adaptation process and enables effective domain adaptive hash learning.
- 7
- AD-AUG: Adversarial Data Augmentation for Counterfactual Recommendation7
A novel counter-factual data augmentation framework AD-AUG to mitigate the impact of the imperfect training data and empower CF models, and design two adversarial learning-based methods from both “bottom-up" data-oriented and “top-down" model-oriented perspectives for counterfactual learning.
- Global Perspectives on the Role of Two-Year/Technical/Junior Colleges in Computing Education7
This panel presents varying global perspectives on the role of community colleges and 2- or 3-year technical schools (collectively called junior colleges here) in computing education.
- Undergraduate IT education in China7
This article systematically introduces recent developments of IT undergraduate education in China and describes a general picture of IT programs at national key universities such as Peking University and Tsinghua University, including core courses, curricula, and knowledge systems.
- Measuring Strength of Ties in Social Network7
It is found that comparing to other types of triads, the one with three significantly strong ties are more likely to be created, which verifies the theory of sociology.
- 6
- 6
- 6
- IT2017 Report6
The novelty of the report is its focus on industry-informed competencies that IT graduates should have to meet the growing demands of a changing technological world in the next decade.
- A Framework for Analyzing Abnormal Emergence in Service Ecosystems Through LLM-Based Agent Intention Mining5
EAMI first employs a dual-perspective thought track mechanism, where an Inspector Agent and an Analysis Agent extract agent intentions under bounded and perfect rationality, followed by a Intention Temporal Emergence diagram for dynamic analysis.
- 5
- Predictive Multi-level Patient Representations from Electronic Health Records5
A Multilevel Representation Model (MRM), which first uses a sparse attention mechanism to model the short-term co-occurrence, then uses interval-based event pooling to remove redundant information and reduce sequence length and finally predicts clinical outcomes through Long Short-Term Memory (LSTM).
- A Context-Based Framework and Method for Learning Object Description and Search5
The proposed work aims to support learners in using massive learning resources from the web to perform self-learning with or without the help of educators' advice and instruction.
- 5
- 4
- 4
- 4
- Personalized Recommendation Algorithm on Microblogs4
A novel recommendation algorithm based on Bayesian personalized ranking (BPR) by modeling user's implicit feedbacks in microblogging community is proposed and experiments show that the recommendation algorithm outperforms all the baselines.
- 4
- 3
- Advances in Medical Imaging Technologies and Their Impact on Clinical Practices3
This comprehensive review explores the latest advancements in medical imaging technologies, including digital X-ray imaging, magnetic resonance imaging, magnetic resonance imaging (MRI), computed tomography (CT), and ultrasound imaging, and their impact on clinical practices.
- 3
- 3
- 3
- Personalized search in digital libraries via spreading activation model3
A novel approach to provide personalized search by learning users' interests from historic behaviors and re-ranking search results by a Spreading Activation (SA) model, which runs on the domain ontology incorporated with a newly defined relationship borrowIntent derived from the assumption of collaborative filtering.
- 3
- 2
- Deep Graph Clustering with Disentangled Representation Learning2
A novel deep graph clustering framework named DisenCluster is proposed, which learns disentangled representations to simultaneously consider node separation results from diverse perspectives and is formulated as a continuous relaxation of the high-order graph cut problem and optimize the objective to obtain effective cluster assignments.
- Future Matters for Present: Towards Effective Physical Simulation over Meshes2
Through pyramid-like alternative propagation between the foresight step and refinement step, the proposed framework FAIR can generate accurate long-term trajectories, achieving a significant error reduction compared with the best baseline on four benchmark datasets.
- A foundation model for bioactivity prediction using pairwise meta-learning2
ActFound, a foundation model for bioactivity prediction trained on 2.3 million experimentally-measured bioactivity compounds and 50, 869 assays from ChEMBL and BindingDB, employs pairwise learning to learn the relative value differences between two compounds within the same assay to circumvent the incompatibility among assays.
- 2
- Computing Competencies and the CC2020 Project2
This birds-of-a-feather session discusses the global project Computing Curricula 2020, a joint project launched by professional computing societies worldwide that examines the current state of curricular guidelines for academic programs granting bachelor's degrees in computing.
- 2
- Should IT2008 be revised?2
In this panel the group members will report on its current progress and will solicit input from participants on what aspects of the model curriculum need to be revisited.
- 2
- Application of PPE model based on accelerating genetic algorithm in the evaluation of slope stability2
The practical example shows that the PPE model is effective and universal and can be used to evaluate other problems and can optimize the course of the PP method.
- KEGOD: Kernel-enhanced Latent Substructure Learning for Graph Out-Of-Distribution Detection1
A novel kernel-enhanced graph substructure learning framework termed KEGOD is proposed for unsupervised graph OOD detection, which captures complementary ID structural semantics in both implicit and explicit manners, enabling reliable detection of OOD graphs through semantic inconsistency.
- A multi-agent system for spine MRI report generation from multi-sequence imaging1
By decomposing report generation into clinically grounded subtasks addressed by specialized agents, the SpineAgent framework enables accurate, interpretable and generalizable spine MRI reporting across diverse imaging sequences and anatomical regions.
- Space-based Parameter Evolving with Lightweight Optimization for Graph Adaptation to Evolving Shifts1
The proposed STEM (State-based Parameter Evolving with Lightweight Optimization), a replay-free framework that transforms continual adaptation into controlled parameter space evolution via a controller-worker architecture, achieves state-of-the-art performance while maintaining minimal computational overhead.
- 1
- 1
- Digital Sobriety: Sustainable Use of Gen AI in Higher Computing Education1
It is argued that Gen AI and its unconscious and enthusiastic adoption expose us as educators to accusations of profligacy in the authors' actions and blind ignorance oof the environmental costs of their actions.
- SPOT: Spectral Optimal Transport for Graph Domain Generalization1
This work formulate the structural augmentation as an optimal transport problem to retain low-frequency key knowledge and solve the problem using Sinkhorn-Knopp algorithm, and incorporates an adaptive perturbation strategy to deep features, where the direction of the additive noise is determined by the homophily degrees to maintain semantic properties.
- Evidential Self-Supervised Graph Representation Learning via Prototype-based Consistency1
This paper proposes a novel framework that employs a graph encoder to capture both local and global information from the original graph and its augmented versions produced through graph diffusion techniques, and constructs learnable prototypes to cover the entire semantic space.
- 1
- 1
- Online Learning Preparation of College Students in Higher Vocational Education1
The study found that teachers were competent and relatively well prepared for the requirements of blended learning, but the actual implementation was less optimistic due to practical factors such as the difficulty of the subject, lack of time, and technical difficulties, and some teachers continued to teach using traditional lecture methods.
- 1
- Let Other Users Help You Find Answers: A Collaborative Question-Answering Method with Continuous Markov Chain Model1
A Collaborative Ranking (ColRank) algorithm employing the Continuous Markov Chain Model (CMCM) to combine the quality of QA pairs and relationships among them is proposed and empirical results show that the innovative algorithm is effective and outperform the state of art Question-Answering baselines.
- 1
- 1
- 1
- 1
- WebGIS-RBDL – A Rare Book Digital Library Supporting Spatio-Temporary Retrieval1
The efforts show that using GIS in digital libraries is a new and promising method to build more effective user-friendly interfaces in digital Library Services, and efficient tools to extract spatio-temporal information automatically from rare-book metadata are developed.
- 1
- Design and application of classroom instruction information pro cessing system (abstract)1
A Classroom Instruction Information Processing System (CIS) is deseribed, consisting of response units, an N-terminal Interface, a real time processor, a multi-function Image/Voice Transfer Unit, a video monitor and a printer.
- Long-Tailed Recognition of Evidential Experts for Graph-level Classification–
This paper develops a principled framework called GraphEVER for long-tailed graph-level classification and proposes an evidence-based routing mechanism to dynamically assign experts, such that the tail classes can receive more attention, while the head classes can reduce redundant engaged experts, further cutting down the computational cost and improving the efficiency.
- –
- –
- PROBE: VLM-Guided Discrete Structural Reconfiguration for Customized and Efficient Video Retrieval–
An efficient VLM-guided video retrieval paradigm PROBE is proposed that leverages a frozen vision–language model (VLM) to inject rich semantic structure into a reusable hash index and introduces criterion-induced subspace regularization that transfers semantic geometry into criterion-conditioned hash subspaces while preserving local relational consistency.
- –
- Architecture-AI Curriculum Integration: Theoretical Frameworks and Pedagogical Innovations–
This study focuses on the interdisciplinary curriculum integration mechanism between archi tecture and artificial intelligence, proposing the “Three-stage Five-chain” teaching system aimed at breaking down disciplinary barriers, constructing a modular connection framework, and promoting the transformation of architect ural education toward intelligence and systematization.
- –
- –
- CODE: Towards Partial Label Graph Learning via Coupled Dual Separation–
This work studies partial label graph learning, in which each graph is linked to a set of candidate labels but only one of them is accurate, and presents a novel approach called Coupled Dual Separation, which contains a message passing branch and a graph kernel branch, which explore graph semantics implicitly and explicitly, respectively.
- An Examination of Computer Graphics and Image Technology’s Role in the Components of Visual Communication Design–
The experimental results show that the modified BCE With Logits visual communication design model can be applied to four different types of images with different levels of tedium, and that when the weight parameter is set to 0.34, the accuracy of image recognition rises to 89.59\%, resulting in suc-cessful visual communication.
- –
- –
- –
- –
- Global Awareness for Computing Educators and Scholars–
This discussion provides participants an opportunity to share global differences in vocabulary and interpretation of idioms and assumptions that pervade teachers and researchers, and to heighten awareness of what changes might lead to a truly global computing education community.
- –
- Hybrid small class teaching–
Hybrid small class teaching, which combines large class lectures with small class recitation, to increase the involvement of students and improve the performance of teaching is presented.
- –
- –
- Teaching of Pharmaceutical Physical Chemistry–
Around the features of Pharmacy, the physical chemistry teaching contents were optimized, the teaching method was improved, the interest of students were mobilized, experiment teaching was emphasize, and the quality of teachers were improved.
- Design and Implementation of Half Automated Virtual Experiment Experimental Report System in the Chemistry of the Junior Middle School–
The author proposes the design scheme of half automated virtual experiment experimental report in the chemistry of the junior middle school,ounds the half automated experiment system structure and its functional and structural design, and takes " To produce oxygen on the condition of the lab" as a case to design half automated experimental report.
- –
- –
- Research on Teaching System of Computer Professional English for the Computer Specialty in University–
Bilingual education based on professional curricula is a more effective method and the teaching system of computer professional English and some suggestions are given.
- –
- Case Teaching Exploration and Practice of Chinese Medicine Preparation Analysis–
The practice shows that case teaching can effectively mobilize students' learning initiative and enthusiasm, improve the ability of communication, and cultivate the innovative thinking and promoting the comprehensive quality enhancement.
- –
- –
- –
- Path Selection for Enterprises to Avoid ERP Black Hole–
Five factors account for ERP black hole, such as ambiguous positioning on ERP, inappropriate ERP product, insufficient support from the top management, conflict with original management mode and over-customized ERP.
- Steam turbine fault analysis based on rule and ontology–
The ontology-based illation method is used to express the fuzzy know-ledge with SPARQL language in the searching form of SQL sentences and to deduce the fuzzy knowledge with Jena Reasoner.
- Implement Web Learning System Based on Genetic Algorithm and Pervasive Agent Ontology–
The information integration method of Semantic Web based on Pervasive Agent Ontology (SWPAO method), which will integrate, analyze and process enormous web information and extract answers for students on the basis of semantics is put forward.
- –
- –
- –
- –
- On the Contemporary Loss and Reconstruction of Service Theory in the Library of Our Country–
Reconstructing service theory should get rid of excessive chains of service technology, means and projects and it should explore the line of sights to search the source of service theory to make the library have its original meaning and the contemporary value under the information technology environment.
- –
- –
- Discussion on the Construction of Electronic Reading Room in Academic Library–
By exploring the basic positions and functions of electronic reading room in academic library, the essay states the basic ways to improve the service and to help student to make full use of the electronicReading room.
- –
- –
- Renewal of literature retrieval teaching niodel for postgraduates and its practice–
The prerequisite for creating a new teaching model, especially the methods such as heuristics, problem solving, discussion and communication, are discussed.
- –
- –
- –
- An exploration about web design and making–
This paper explores a technology and method of web design and making from HTML language, layout design, which makes different aspects of a matter, as well as takes notice of question.
- –
- –
- University library administration under network environment–
The administration of humanization, in for-mation collection and subsystem of information man resources, and resource-sharing, etc in the university libraries under network environment is discussed.
- Improving the Accuracy of Long Jump Run-up–
This paper analyzes and studies the random test data, which 67 students of PE department got, and finds out the principal factors that affect the accuracy of long jump run up so as to improve the teaching methods and the teaching quality.
- Education ideas and models of administrative structure teaching–
Administrative structure teaching includes structure theory and education ideasl Teaching models such as researching ones; problems discovering ones; and many other teaching skills and methods can be derived from it on the basis of class teaching model.
- –
- –
- –
- Creating an environment of modern educational technology–
Motivated teachers must take an active part in applying modern educational technology, exploring new educational model, implementing quality education and contributing to the modernization of the authors' education in the new century.
- –
- –
Publication data from OpenAlex, with missing venues and authors filled in from Crossref; citation counts are the higher of OpenAlex and Semantic Scholar, last synced 2026-10-15. One-sentence summaries under some papers are written by Semantic Scholar’s model. Citation counts may be lower than on Google Scholar, which indexes more sources.
Report an error
Wrong papers, two people merged into one, or a profile that should not be here? Tell us and we will fix or hide it.